Siemens, Nvidia, and UK robotics startup Humanoid have successfully deployed an AI-powered wheeled humanoid robot in live logistics operations at a Siemens electronics factory in Germany.
The HMND 01 Alpha completed over eight hours of autonomous tote-handling at 60 moves per hour with a pick-and-place success rate above 90%, and was integrated directly into Siemens’ production systems.
Siemens and UK robotics company Humanoid, in partnership with Nvidia, have announced the successful deployment of an AI-powered humanoid robot in live logistics operations at Siemens’ electronics factory in Erlangen, Germany.
The robot, Humanoid’s HMND 01 Alpha wheeled model, built on Nvidia’s physical AI stack, autonomously handled tote-destacking tasks for over eight hours, reaching a throughput of 60 container moves per hour and a pick-and-place success rate above 90%.
The announcement was made at Hannover Messe 2026, building on a Siemens–Nvidia strategic partnership first announced at CES.
The 💜 of EU tech
The latest rumblings from the EU tech scene, a story from our wise ol’ founder Boris, and some questionable AI art. It’s free, every week, in your inbox. Sign up now!
The task itself was unglamorous by design: picking totes from storage stacks, transporting them to conveyor belts, and placing them at designated pickup points for human workers.
Repetitive, physically demanding, and exactly the kind of work that industrial automation has historically struggled to handle when the environment is unpredictable, the objects are inconsistently placed, or the task requires coordinating with humans in real time.
The Erlangen trial is significant precisely because it ran in a live production environment, not a controlled lab, alongside human operators and other automated systems, with real production consequences if the robot failed.
The HMND 01 Alpha combines a wheeled lower platform with a humanoid upper body equipped with advanced manipulation capabilities.
Integration into Siemens’ factory was handled through the Siemens Xcelerator platform, which provides a digital twin, AI-enabled perception, PLC-robot interfaces, fleet management, and industrial communication networks.
This allowed the robot to coordinate in real time with production systems, other autonomous guided vehicles, and human workers, the kind of deep integration that the companies argue separates a genuine factory deployment from a showpiece demonstration.
Stephan Schlauss, Global Head of Manufacturing Motion Control at Siemens, described the Erlangen plant as “customer zero”, meaning Siemens used its own factory as the test ground before offering the capability to customers.
On the Nvidia side, the HMND 01 Alpha uses Nvidia Jetson Thor for edge compute, Nvidia Isaac Sim for simulation, and Nvidia Isaac Lab for reinforcement learning and policy training.
The simulation-first development approach, training and validating the robot’s behaviours in a virtual environment before physical deployment, is what allowed Humanoid to compress prototype development from the industry-typical 18 to 24 months down to approximately seven months, the companies said.
Deepu Talla, Nvidia’s vice president of robotics and edge AI, described the deployment as “paving the way for humanoid robots meeting real production targets on a live factory floor.”
Humanoid, founded in 2024 by Artem Sokolov, is based in London with offices in Boston and Vancouver, and brings together over 200 engineers from technology companies. The company also makes a bipedal version of the HMND 01 Alpha, which has 29 degrees of freedom and is equipped with RGB cameras, depth sensors, and 6D force/torque sensors.
The wheeled model used in Erlangen has previously been tested in a proof-of-concept with Schaeffler for picking metallic bearing rings. The Siemens trial, which ran for two weeks in January 2026 ahead of the April announcement, was the most demanding deployment to date.
The companies were careful not to overstate their timelines. They described the Erlangen trial as “a milestone in the journey to bring physical AI from vision to industrial reality” but did not provide a roadmap for commercial rollout.
The broader significance, as Siemens frames it, is the establishment of a “factory-grade model” for humanoid deployment that other companies can replicate, a reference architecture rather than a one-off.
The partnership sits within a wider industrial trend: humanoid robots capable of working in human-centred environments are increasingly being positioned as the solution to labour shortages in manufacturing sectors where fully automated lines are impractical, either due to product variability, safety constraints, or the need for human-robot collaboration.
